worktrees

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill defines a protocol for using Git commands to manage worktrees (e.g., git worktree add, git worktree remove). The instructions emphasize a safety-first approach, requiring the agent to perform pre-flight checks and obtain explicit user confirmation before executing any destructive or significant repository mutations.
  • [INDIRECT_PROMPT_INJECTION]: The workflow processes user-provided or project-derived variables like <slug> and <branch-name> and interpolates them into shell commands, creating a potential surface for indirect injection.
  • Ingestion points: Path slugs and branch names derived from user requests or file system structure.
  • Boundary markers: No explicit delimiters are used to wrap these variables in the shell command templates.
  • Capability inventory: The skill utilizes shell-based Git commands which have access to the local repository and file system.
  • Sanitization: While no automated sanitization is described, the manual user confirmation requirement for the exact command string serves as a critical security control to prevent unauthorized command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 07:22 AM
Security Audit — agent-trust-hub — worktrees